All bleeding stops — but does idarucizumab (Praxbind) make it stop faster?

July 27, 2016, 2:56 pm


Persistent life-threatening hemorrhage after administration of idarucizumab. Alhashem HM et al. Am J Emerg Med 2016 June 30 [Epub ahead of print]


Dabigatran (Pradaxa) is a direct thrombin inhibitor approved for stroke and embolism prophylaxis in patients with non-valve-related atrial fibrillation. When it was first released in 2008, a major disincentive to widespread use was the lack of a reliable reversal agent to treat major bleeds, or to administer before necessary invasive procedures.

In October 2015, the U.S. Food and Drug Administration approved idarucizumab (Praxbind), a monoclonal antibody that avidly binds to dabigatran. under its accelerated approval program. As described by an FDA release, this program:

. . . allowed drugs for serious conditions that filled an unmet medical need to be approved base on a surrogate endpoint.”

In the case of idarucizumab, the surrogate end-points involved normalization of laboratory parameters of anticoagulation. As far as I can determine, there have been no studies that demonstrate convincingly any clinical patient-oriented benefit. In fact, in the major study addressing this issue,  it took a median of 11.4 hours to restore hemostatis after administration of idarucizumab. There was not control group, so we have no idea if this is better than simple watchful waiting. Clearly, it seems far too long to be useful in true life-threatening hemorrhage.

This case report illustrates the point.

A 65-year-old man recently started on dabigatran for atrial fibrillation presented to the emergency department weakness and dyspnea that started approximately 1 hours before. He gave a history of melena for 3 days. On arrival he ws tachycardia (122 ppm) and hypotensive (BP 74/52 mmHg.)

Digital rectal examination confirmed the presence of melena, and a nasogastric tube returned 300 ml bright red blood that did not clear with irrigation. The patient remained unstable despite administration of fluids and packed red blood cells.

The authors note that a markedly elevated thrombin time (120 sec, reference 15-19 sec) indicated significant dabigatran activity. After idarucizumab (5 gm) was administered intravenously. the coagulation tests improved but the patient remained unstable. Upper endoscopy revealed ongoing hemorrhage from a vessel in the duodenum. Several attempts to control the site of bleeding failed, and the patient ultimately underwent angiography with embolization of the gastroduodenal artery. This successfully stopped the bleeding and the patient was discharged after a 4-day hospital stay.

According to Wikipedia, the hospital acquisition cost of a 5 gm dose of idarucizumab is $3500. Although it clearly improves coagulation lab values, it is not clear if if stops significant or life-threatening hemorrhage, or allows for safer procedures such as hemodialysis.

An excellent post at REBEL EM discussed many of the problems with the major study of idarucizumab mentioned above, including industry sponsorship, lack of power, dicey inclusion criteria, and poor study design without blinding or randomization.

Bottom line: idarucizumab had accelerated approval by the FDA as a reversal agent for dabigatran because of biological plausibility. There is still no real proof of its effectiveness. However, in cases of truly life-threatening hemorrhage, many clinicians will no doubt administer the drug if no other treatment options seem available.

Related points:

Case report: hemodialysis for dabigatran overdose

The many potential problems with using dabigatran

Case series: four patients with dabigatran-associated bleeding

Review: the bleeding patient on dabigatram

Dabigatran and the elderly






  1. ABCIVO2monitor Says:

    Agree that the use of surrogate endpoints in the idarucizumab literature is worrisome, and that the (currently ongoing) REVERSE-AD trial certainly suffers from poor design. However, I’m not sure if this particular case makes a strong argument against the efficacy of idarucizumab. Even in a non-anticoagulated patient, a significant GI arterial bleed will require VIR/GI/surgical intervention for hemorrhage control.

  2. Leon Gussow Says:


    I agree that this one case report taken in isolation is not dispositive. It did serve to remind me of the key point that in the only major study of the drug, median time to restoration of hemostasis was 11+ hours.